Abstract 114: Mechanical Circulatory Support in Percutaneous Coronary Intervention: A Propensity-matched Analysis

Abstract

Introduction: Percutaneous ventricular assist devices (PVAD) and intra-aortic balloon pump (IABP) are used to provide mechanical circulatory support for high-risk percutaneous coronary intervention (PCI). Due to limited evidence from randomized controlled trials, we compared clinical outcomes between PVAD and IABP in PCI patients using a propensity-matched analysis. Methods: Adult patients who underwent PCI during 2004-2012 and also received either a PVAD or an IABP on the same day as PCI were identified in the National Inpatient Sample using ICD9 procedure codes. We compared in-hospital mortality for PVAD vs. IABP using a 1:2 propensity-matched analysis - both overall and in subgroups with cardiogenic shock, AMI without cardiogenic shock and no cardiogenic shock or AMI. Results: We identified 5031 patients who received a PVAD and 122,333 who received an IABP on the same day as PCI. Patients who received PVAD were older (69 vs 65 years), more likely to be men (74% vs 69%), admitted electively (30% vs 11%) but less likely to have AMI (52% vs 90%), cardiogenic shock (23% vs 50%), cardiac arrest (12% vs 25%) or need mechanical ventilation (16% vs 29%) compared to IABP patients (P<0.001 for all). In contrast, prevalence of heart failure (68% vs 41%), valvular heart disease (22% vs 13%), chronic kidney disease (27% vs 11%), hypertension (71% vs 56%) and diabetes (46% vs 32%) was higher in PVAD recipients (P<0.001 for all). Unadjusted in-hospital mortality in PVAD recipients was lower compared to IABP patients - both overall (12.8% vs 20.9%, P<0.001) and in the cardiogenic shock subgroup (31% vs 38%, P=0.04) but was similar in patients without cardiogenic shock. After propensity-matching and successful balancing of covariates (figure) we found no difference in mortality in PVAD and IABP recipients (odds ratio [OR] 0.88, 95% CI 0.70-1.09). Our findings were also consistent among patients with cardiogenic shock (OR 1.37, 95% CI 0.99-1.90), AMI without cardiogenic shock (OR 0.72, 95% CI 0.46-1.14) and no cardiogenic shock or AMI (OR 0.54, 95% CI 0.27-1.06). Conclusion: The lower unadjusted mortality in patients undergoing PCI with PVAD support compared to IABP support may be due to selective use of PVADs in a lower risk population. Randomized trials are necessary to establish the clinical effectiveness of PVADs to support high-risk PCI.